A Design Proposal
For Accountability Under ESSA
By Teach Plus Teaching Policy Fellows

As a group of high-performing teachers who teach in high-poverty schools, we have learned from our experiences in classrooms across America that the learning gaps among subgroups of students are not the result of differences in the abilities or talents of students, but rather the result of a broken public education system—with differences in expectations, access to effective teachers, access to purposeful school cultures, and access to enriching learning opportunities. These differences are uniformly tilted to the disadvantage of the students who fill our classrooms: students from low-income families and children of color. Our experiences drive the convictions that underlie our design for accountability systems under ESSA.

Our Design Principles

We propose a two-tiered accountability system. This layered approach allows for both reporting and learning which factors are most important to school success so that states can best direct resources and support to schools that need it the most. The second tier will provide metrics that are informative but not determinative. The rationale for such a tiered system is based on several design principles:

• We need to debate what is most important in determining a school’s quality and then drive resources toward what is most important and will make the most difference for kids.
• The rating system needs to be simple, easy to understand, and easy to act upon. The alternative, an overly complex index of multiple measures, can undermine transparency and make it difficult to interpret numbers and design next steps.
• Creating rating systems based on too many metrics will water down the significance and weight of these most important factors.
• While Tier 1 indicators measure the most important outcomes, schools and districts need to be able to make evidence-based decisions for ways to improve. By analyzing the impact of shifts in Tier 2 indicators on Tier 1 results, schools and districts will be able to form actionable hypotheses on how best to drive improvement.

TIER 1 INDICATORS

Tier 1 Indicators are what we deem to be the most important performance indicators for school quality. Ratings in our accountability system will be based on a composite score of a school’s performance across these Indicators.

Tier 1 Academic Indicators of School Quality

1. Measure of Grade Level Proficiency: Using assessment data from annual statewide assessments (PARCC, SBAC or other statewide assessment)

• Our design places a primary focus on student achievement in reading and math. At all times, we need to ensure that students are learning and achieving at high levels.
• Four Performance Levels: Exceeds Standard, At/Near Standard, Below Standard, Far Below Standard
• Weighting for Scores: 1.25x for Exceeds Standard. 1.0x for At/Near Standard, .75x for Below Standard, or .50x for Far Below Standard
• This weighting system incentivizes moving students up levels. Schools will be incentivized to move students up levels to earn higher ratings.
• All student scores will be added after weighting for performance level. The total student score after weighting will be divided over total possible points.

2. Measure of Progress Toward English Language Learner (ELL) Proficiency

• Progress towards English Language Proficiency would be gauged by two measures: ACCESS scores (weighted for both time to proficiency and growth), and metrics that support student transition from ESOL to full inclusion in general education. While ACCESS scores can be looked at three ways (raw scores, scaled scores, and proficiency scores), progress should be determined by looking at the scaled scores. While all students are expected to grow, students who are just beginning the program would be expected to have more growth than students who have been in it for a longer period of time.

3. Measure of Student Growth: Using assessment data from NWEA or another norm-referenced, computer-adaptive test that can assess knowledge and skills tri-annually over spans of grades

• While proficiency is important, we must recognize and reward growth in students and cohorts. This would be done using assessment data from NWEA or another norm-referenced, computer-adaptive test that can assess knowledge and skills over spans of grades. Reading assessments should include an equal ratio of fiction and non-fiction passages. Other assessments should be limited and would not be included in the school ranking.
• Testing Time: Students will be tested in fall, winter and spring. Testing time for math and reading for each part of the year will last no longer than an average of ninety minutes per test. We need to steadfastly guard against too much testing time.
• Calculating Student Growth: Baseline scores for schools will be determined by the number of students scoring Proficient. Using the average percentile of the last three recorded scores, the state will create an Expected Magnitude of Growth between the average and the student’s spring scores. After calculating an Expected Magnitude of Growth for each student, schools will calculate an Actual Magnitude of Growth for each student. We will then average the actual magnitude of growth vs. expected magnitude of growth to calculate students’ growth over time. The numbers should be weighted several ways:

• Weight 1: Advanced Students: Schools will receive additional points for students scoring Advanced in order to incentivize schools to push students who are already secure on grade level standards and because advanced programs are often cut or overlooked in high-poverty/minority schools.
• Weight 2: Later Grades: Scores will be weighted by grade level, so that fifth-grade scores will have more weight than fourth-grade scores, which will have greater weight than third-grade scores. Such a system incentivizes schools to address chronically low-performing students and encourages vertical collaboration.
• Weight 3: Basic Students: Schools will receive partial credit for students scoring Basic. We believe this would incentivize schools to focus on moving students who are significantly below grade level to the next band of achievement. The number of students scoring Basic is a useful indicator of progress at the school level.
• Weight 4: Mobile Students: Our model will account for high rates of mobility in student populations because student scores will be weighted to count more when they have been at the school longer. For instance, a fifth grader who has been at the school since kindergarten should could more than a fourth grader who just arrived at the school in November of that year.

Tier 1 Non-Academic Indicators of School Quality

Schools that are welcoming, safe, and learning-focused produce the strongest academic results. States will assess this through the Tier 1 indicators of attendance rates and an annual survey of students, staff, and families.

1. School Climate and Culture: Data provided through a survey of students, staff, and families.

• The survey will be based on the 5 Essentials System, developed by the University of Chicago Consortium on School Research as an evidence-based approach to measuring a school’s strengths and weaknesses in climate and culture. The survey would measure school leadership, teacher collaboration, family engagement, school environment, and ambitious instruction. Schools will receive scores in the areas of effective leadership, teacher collaboration, family engagement, supportive environment, and ambitious instruction; they will then set annual goals around each of the five areas and create action plans as needed.

2. Student Attendance: Overall in-seat school attendance rates.

• Attendance will be tracked according to district policies.
• Schools will need to maintain 95 percent student attendance. Students who aren’t in school aren’t learning. Schools will also have to identify chronic absenteeism, which would be students who have more than ten absences a semester.
• Schools that do not meet the 95 percent requirement will create action plans in order to increase student attendance. Schools with a high percentage of chronically absent students would need to include addressing that issue in their action plan.

TIER 2 INDICATORS

The purpose of the Tier 2 Indicators is to provide for comprehensive reporting of a variety of other important factors of school quality that will help drive the interventions we believe are necessary to help schools meet Tier 1 Indicators.

Tier 2 Indicators provide robust information on a variety of potentially predictive inputs on school quality. States, districts, and schools will be able to analyze and study their performance on these Indicators and examine whether certain factors have predictive power in learning to improve school and district performance on the Tier 1 factors. In this way, states will have the opportunity to identify struggling schools or subgroups and form actionable hypotheses for improving outcomes, using disciplined inquiry to drive improvement.

States will set short-term and long-term goals for Tier 2 Indicators. Instead of including them in the rating system, states will use “traffic lighting" to highlight which Tier 2 Indicators meet important goals and which do not.

Tier 2 Academic Indicators

1. Assessments in science, civics, and additional subjects, as decided by the state, besides ELA and mathematics.
2. STEAM programs or integrated curriculum
3. English language learner integration, exiting from ELL status.
4. Student mobility across levels over time, increased weighting for later grades in a school.

Tier 2 Non-Academic Indicators of School Quality

These secondary measures of school quality will not be measured by the state, but should be gauged by schools as they reflect on their strengths and growth areas. They could include the following:

• School Climate and Culture Indicators: Teacher leadership opportunities, teacher mentoring programs, advanced courses, anti-bullying campaigns, peer mediation programs, peer tutoring programs, celebrations of academic success, arts-integrated curriculum, music ensembles, after-school clubs, parent-led workshops, sports teams, elective courses, and library and computer lab availability.
• Attendance Indicators: Interventions that have been shown to positively affect attendance rates include the following: a proactive home visit program, recognition for improved attendance, enforcement of truancy laws, and enlisting community members.
• Wrap-Around Services: Schools should offer supports for students in families to increase access for families to community services. This shouldn’t be limited to counselors and social workers, but could also include opportunities to increase family involvement at schools.

Using Data to Drive Resource Allocation

Assessment data must be disaggregated by subgroup and reported to Local Education Agencies (LEAs), parents, and the public. If gaps exist, LEAs must submit a plan for how to use school- or district-identified interventions to reduce inequities and increase success for identified subgroups. The statewide accountability system would include auditing powers to make sure LEAs are using the resources and funds either for the subgroups that are underperforming or for maintaining growth. This twofold approach focuses on capacity building and is targeted to ensure resources are used to improve student outcomes (i.e., success rates of underperforming subgroups and maintaining their growth.)